JPH02202175A - Sharpness emphasizing method for picture scanning and recording device - Google Patents

Abstract

PURPOSE:To obtain a visually preferable output picture with high sharpness by making the spatial frequency component of a picture signal to be emphasized variable in correspondence to the size of the output picture. CONSTITUTION:At first, the size of the output picture is obtained based on a designated condition and next, a standard observation distance is estimated in correspondence to the size of the output picture. Then, the space frequency component of the necessary picture signal is obtained to strongly contributed to the sharpness when the output picture is observed from the estimated standard observation distance. After that, the obtained spatial frequency component is emphasized. Thus, by setting the width of edge emphasis corresponding to the size of the output picture, the visually preferable copied picture with high sharpness can be obtained.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention relates to a sharpness enhancement method in an image scanning recording device such as a printing plate-making scanner or a gloss shimiri, and in particular,
Concerning how to set the optimal edge enhancement width.

<Prior art> The so-called sharpness, which refers to the sense of how sharp the contours (edges) of an image are, is a measure of image quality,
In particular, it is one of the important parameters that influences the visual desirability of an image. For example, when viewing an image, it is known that the spatial frequency component around 12 cpd in the image strongly contributes to the sharpness. (Breakfast Shoten Image Engineering Handbook P25), here, c p d (cyele
per degree) refers to the number of bears of white and black patterns that exist within a viewing angle of 1 degree.Therefore, the spatial frequency components around 12 cpd in an image are the spatial frequency components of the image signal that are This refers to a spatial frequency component that allows approximately 12 black and white patterns to be created within a viewing angle of 1 degree.

Note that various units are used to express spatial frequency, but in this specification, the above-mentioned cpd, 1ρ/scale, (
Linebear/llll1) is used. In addition, to prevent terminology confusion, the spatial frequency as seen from the viewer, expressed in cpd, is expressed as "apparent spatial frequency," and lp/
The spatial frequency on the duplicate image displayed in am is simply referred to as "spatial frequency."

Naturally, cpd and ip/-1 can be converted into each other if viewing distance is given.

When duplicating an image, by appropriately emphasizing the apparent spatial frequency component around 12 cpd, an appropriate width of edge enhancement can be obtained, and a visually pleasing duplicated image with high sharpness can be obtained.

Incidentally, the division distance on an image divided by a viewing angle of 1 degree increases as the distance between the image and the viewer increases. Therefore, the frequency components for creating approximately 12 black and white patterns in this section, that is, the spatial frequency components on the replicated image corresponding to an apparent spatial frequency of 12 cpd, are determined by the distance between the image and the viewer. As the spread of
However, when copying originals and making printed matter, it is necessary to specify the spatial frequency component that is the target of sharpness enhancement. The distance to the viewer is the distance of clear vision M (generally,
25~30c+m), and 12cp at this time
Usually, the spatial frequency component corresponding to the vicinity of d is emphasized.

Of course, sharpness is a visual preference, so the detailed parameters that determine how to apply sharpness enhancement depend on the device that applies sharpness enhancement (for example, a prepress scanner)
For example, some devices vary the spatial frequency components to be emphasized depending on the replication magnification. Sharpness also depends on the preference of the operator using the device.
There are some things that change. However, the average numerical value of the spatial frequency components to be emphasized can be summarized as described above.

<Problems to be Solved by the Invention> However, the conventional sharpness enhancement method described above has the following problems.

In other words, in reality, not all types of printed materials are viewed at clear viewing distances. Since the image is normally viewed visually, high sharpness can be obtained by emphasizing the spatial frequency component corresponding to around 12 cpd at the distance of clear vision.

However, printed materials such as large-sized posters are usually viewed from a distance greater than the distance for clear vision. Even for such large-sized printed matter, the distance of clear vision (for example, 2
For example, if the spatial frequency component corresponding to around 12 cpd when viewed at a distance of 1 m is reproduced, the printed matter will be reproduced at 48 cpd when viewed from a distance of 1 m.
This means that spatial frequency components corresponding to the vicinity are emphasized. In other words, from this distance, the width of edge enhancement is too narrow and high sharpness cannot be obtained.The same thing can be done by simply emphasizing according to the replication magnification without considering the size of the output image. The same can be said of conventional methods in which the power spatial frequency component is varied.

In this way, conventional methods emphasize specific spatial frequency components in the image signal regardless of the size of the output image, so if the viewing distance differs depending on the size of the output image, However, there is a problem in that a sufficient sharpness enhancement effect cannot be obtained.

The present invention has been made to solve these problems, and by setting the width of edge enhancement according to the size of the output image, it is possible to create a visually pleasing reproduced image with high sharpness. The purpose of this paper is to provide a sharpness enhancement method that can obtain the following.

<Means and operations for solving i*a> The configuration and operations of the sharpness enhancement method according to the present invention will be explained in order below.

(1) Find the size of the output image based on specified conditions.

For example, in a prepress scanner, when recording an image, the size of the document, the trimming range in the document, the reproduction magnification, etc. are specified in advance.In the present invention, first, the output image is Find the size. of course,
The method of specifying the conditions is arbitrary; for example, the size of the output image may be directly specified.

[2] Estimate the standard viewing distance according to the size of the output image.

When fixating an image, many people unconsciously take actions such as moving closer to a small image and fixating it, and moving away from a larger image. Although such behavior may be unconscious, it is thought that it contains a certain regularity. In the present invention, such a regularity is determined in an appropriate manner, and a preferable viewing distance according to the size of the output image is estimated from this regularity, and this is used as a standard viewing distance.

An example of a specific method for estimating the standard viewing distance will be described below.

For example, the TV Society Journal (88.7 Lecture HDTV Part 2)
According to a report from 2013, when viewing images with sufficiently high resolution, a viewing distance of 2 to 3 H (H: screen height, angle of view of 20 to 30 degrees) is most preferred. If it is less than this, the feeling of pressure from the screen is strong, and the entire screen cannot be seen at once, so the preference becomes less favorable.

This is the result of an experiment conducted on a high-definition television (HDTV) screen, so the screen's aspect ratio is constant at about 3:5. In this experiment, the viewing distance was 2H.
In this case, the vertical angle of view is approximately 30 degrees, and the horizontal angle of view is approximately 50 degrees.

When the above experimental results are applied to a printed matter whose aspect ratio is not constant, the distance at which the angle of view looking into the long side of the printed matter is about 50 degrees can be estimated as the standard viewing distance. For example, if you fixate your eyes from a position where the angle of view looking at the long side of an A4-sized print is 50 degrees, the viewing distance at that time will be about 32 cm, and this tit determination method matches well with everyday experience.

[3] Find the spatial frequency component of the required image signal that strongly contributes to the sharpness when fixating the output image from the estimated standard viewing distance.

Specifically, the spatial frequency component of the image signal to be emphasized is
This is a spatial frequency component corresponding to an apparent spatial frequency of around 12 cpd. This spatial frequency component is calculated as follows. For example, for an A3 size print, the standard viewing distance is estimated to be 45 c-, and the apparent spatial frequency component of 12 cpd when viewed from this distance is:
On printed matter, it is approximately 1.51p per 1an (1ine
pair) corresponds to the spatial frequency component of (see the following formula (■)). Here, 1.51ρ per [■ is 1.51ρ per [■]
This means that there are 1.5 pairs of black and white patterns within the range.

1÷(450Xtanl'÷12) #1.5 1p
/am -■Similarly, in the case of A1 size printed matter, the standard viewing distance is 90cm, and at this time 12cp
The apparent spatial frequency component of d is 1m1m on the printed matter.
This corresponds to a spatial frequency component of approximately 0.761p (the following formula % formula % [4] Emphasize the obtained spatial frequency component.

Methods for enhancing sharpness include an optical analog method and a digital method, and either method can change the spatial frequency component to be emphasized.

(1) Analog method Please refer to Figures 3 to 5 below. Figure 3 is a schematic diagram of the main aperture MA and sub aperture SA used for photoelectric scanning, Figure 4 is a waveform diagram of signals related to sharpness enhancement processing, and the fifth leap represents the spatial frequency characteristics (MTF) of each signal.
It is.

When a document like the one shown in FIG. 4(a) is photoelectrically scanned, the main aperture MA, which is a relatively small aperture,
A sharp signal S (see FIG. 4(b)) is obtained from the sub-aperture SA, which is a relatively large aperture, and an unsharp signal U (see FIG. 3(c)) is obtained from the sub-aperture SA, which is a relatively large aperture. By finding the difference between these two signals and adding the difference signal k(SU) (see FIG. 4(d)), which is increased or decreased as appropriate, to the sharp signal S, the sharpness emphasized signal S+k
It is well known that (SU) (see FIG. 4(e)) can be obtained.

Here, a sharp signal S and an unsharp signal U.

Difference signal k (S-tJ), sharpness emphasis signal S+k (S
-()) are shown in Fig. 5 (), respectively.
It will be as shown in a) to (d). Spatial frequency components that strongly contribute to sharpness are spatial frequency components f, shown in FIG. 5(d),
Therefore, when the output image is fixed at a standard fixation distance, this spatial frequency component may be controlled to have a frequency corresponding to an apparent spatial frequency of 12 cpd.

Specifically, by keeping the size of the main aperture MA constant and changing the size of the sub-aperture SA, the spatial frequency f0 can be varied.For example, in the case of the above-mentioned A3 size printed matter. ,
In order to emphasize the spatial frequency component of 1.51ρ/-, a sub-aperture SA with an effective width of 1/1.5 = 0.66 mm may be used.

In addition, in prepress scanners, it is common to perform magnification conversion processing at the same time, but in this case, by setting the effective size of the sub-aperture SA to the above value x 1/magnification, the necessary Needless to say, the spatial frequency component element 6 can be obtained.

That is, in the case of the upper side, the size of the main aperture MA that samples the original image on the input side is L (main scanning direction) x M (sub scanning direction) (II gull Xam) on the original image, which is the same due to magnification conversion. If the pixel is on the output image, L'
When converted to the size of M', the subaperture SA
The effective size of is 0.66xL/in the main scanning direction.
The width of L' (+u+) in the sub-scanning direction is 0.6
The width may be 6XM/M' (+u+).

(2) Digital method In the digital method, in order to obtain an unsharp signal, image signals within a pixel area in which the required number of pixels are arranged in the main scanning direction and the sub-scanning direction are averaged, as shown in Figure 6. do. That is, the image signal obtained by photoelectrically scanning the original is digitally sampled to obtain an image signal corresponding to each pixel in the pixel area.The center pixel in the 0 pixel area (pixel in the hypotenuse area in the center of FIG. 6) The image signal corresponding to the image signal is defined as a sharp signal, and the image signal obtained by averaging the image signals for all pixels is defined as an unsharp signal. By determining the difference between these two signals and adding this difference signal to the sharp signal, a sharpness emphasis signal is obtained.

Here, by keeping the size of the central pixel related to the sharp signal constant and changing the range of surrounding pixels to be averaged, the spatial frequency component to be emphasized can be changed. In, 1.
When emphasizing the spatial frequency component of 51p/s-, 1/1
．． It is only necessary to average the surrounding pixels in the range of 5 = 0.665 m. Therefore, if the size of one pixel is 50 μm, from 0.6610.05-13.2 #13, 1
It is sufficient to average the 3×13 pixel areas.

<Example> Hereinafter, an example of the present invention will be described with reference to the drawings. 1st
The figure is a block diagram of main parts of an image scanning and recording apparatus that performs digital sharpness enhancement processing.

In the figure, reference numeral 1 is a line memory group for delaying the digitized image signal one scanning line at a time, and Ll-L
It is composed of 15 line memories. Of these, Ll is a line memory containing the center pixel.

2 has seven types of pixel areas depending on the size of the output image,
That is, 3 x 3.5 x 5.7 x 7.9 x 9.11 X 1
This is an averaging processing unit that performs averaging processing on a pixel area of 1.13×13.15×15. The averaging processing unit 2 includes an addition circuit 3 that adds image signals of pixels arranged in the sub-scanning direction in each pixel area, an addition processing unit 4 that adds the output of the addition circuit 3 in the main scanning direction, and an addition processing unit 4 that adds the outputs of the addition circuit 3 in the main scanning direction. It includes a division circuit 7 that divides the output signal of 3 by a required number to obtain the average value of the image signal in each pixel area. ..., 1
It is shown that each of the 5×15 pixel areas is a component that undergoes averaging processing.

Since the averaging process for each pixel area is the same, the averaging process for a 5×5 pixel area will be explained here as an example.

Of the line memory group l, image signals of five pixel groups each in the first to fifth columns arranged in the sub-scanning direction are sequentially applied to the adder circuit 3° and added by the line memories L5 to L9. . These addition signals are processed by the addition processing section 4. Shift register 5. are given in order. Shift register 5. The added signals for the five columns are taken out in parallel,
These are the adder circuit 6. given to. Addition circuit 6. teeth,
An integrated value of image signals for all pixels in the 5×5 pixel area is calculated, and this integrated value is given to the division circuit 7s. Division circuit 7. outputs an unsharp signal, which is an averaged image signal within the 5x5i!ii pixel area, by dividing the integrated value by the number of pixels in the pixel area, 251.

In this way, in the averaging processing section 2, the unsharp signal U1. ..., U
, , are applied to the selection circuit 8, respectively. Selection circuit 8
is based on the control signal from the microcomputer 9,
A desired unsharp signal is selected from among the unsharp signals Us, . . . , USs and output.

The unsharp signal selection control procedure in the microcomputer 9 will be described below with reference to the flowchart shown in FIG.

Step S1: The microcomputer 9 receives the size of the read document from the document size input section 10, the trimming range within the document from the trimming range input section 11, and the copy magnification from the copy magnification input section 12.

The microcomputer 9 calculates the size of the output image based on these specified conditions in step S2; calculates and sets a standard viewing distance estimated according to the calculated size of the output image. As described above, the standard viewing distance is such that the angle of view looking into the long side of the output image is about 50 degrees.

Step S3j A spatial frequency component corresponding to an apparent spatial frequency of 12 cpd when the output image is fixedly viewed from the set standard viewing distance is calculated.

When the JJ1 standard viewing distance is L, the spatial frequency component @ (Ip /5se) corresponding to 12 cpd can be obtained by the following equation (2).

fo = 1 + (LXtanl°+12)
-... ■Step S4: Determine a pixel area suitable for emphasizing the calculated spatial frequency component. For the calculated spatial frequency component, the length of 1p is 1/f
o (arm) From Te Toru, -side length is 1/
f. It is sufficient to select the pixel region closest to the pixel region from among the seven types of pixel regions. For example, if the size of one pixel in each pixel region is 5Q71m, the pixel region is determined as follows.

When i/f, <0.2, 3×3 pixel area 0.2 ≦1/
f, <0.3 (7) when 5×5 pixel area 0.3≦1
/f, when <0.4, 7×7 pixel area 0.4≦1/f*
<0.5, 9×9 pixel area 0.5≦1/1. <0.
6: uxii pixel area: 0.6≦1/f6<0.7: 13 x 13 pixel area: 0.7≦1/f0: 15 x 15 pixel area Note that magnification conversion processing is performed after sharpness enhancement processing. In this case, the frequency of the emphasized component changes due to the magnification conversion process, so it is necessary to take this into consideration in determining the pixel area. , 1/r of the conditional expression
The required pixel area is determined by replacing 0 with (1/fo) x (100/A). When the magnification conversion process is performed before the sharpness enhancement process, the pixel area can be determined using the above conditional expression without considering the duplication magnification.

Step S5 When the two pixel areas are determined, a control signal for selecting the pixel area is output to the selection circuit 8. Thereby, the selection circuit 8 selectively outputs the unsharp signal related to the determined pixel area.

Hereinafter, the unsharp signal selected by the 0 selection circuit 8 shown in FIG. 1 is given as one input to the next-stage difference circuit 13.

On the other hand, the sharp signal S related to the center pixel included in the line memory L7 is provided as the other input to the difference circuit 13 and the addition circuit 16 via the delay circuit 14 whose delay time is controlled by the microcomputer 9. The difference circuit 13 outputs a difference signal S-U between the sharp signal S and the unsharp signal U, and this difference signal is multiplied by an appropriate coefficient k in the next stage multiplier circuit 15 to be increased or decreased, and then added. It is given as one input of circuit 16. The addition circuit 16 adds the sharp signal S and the difference signal k (S-U) and outputs the added signal as a sharpness-enhanced digital image signal S+k (S-U).

As described above, according to this embodiment, an appropriate pixel area is selected according to the size of the output image, and the averaged image signal of that pixel area is used as an unsharp signal. The edge enhancement width is appropriately varied accordingly, and a visually preferable sharpness enhancement is performed in accordance with the size of the output image.

In the above example, the standard fixation distance was determined from the size of the output image, and then the spatial frequency component corresponding to 12 cpd was determined, and then an appropriate pixel area was determined. Create a table in advance that defines the relationship between the image size and the corresponding pixel area,
By referring to this table, an appropriate pixel area may be determined directly from the size of the output image determined from the specified conditions.

Furthermore, when the present invention is applied to optical analog processing, a plurality of sub-apertures of different sizes are prepared according to the size of the output image, and a selection signal from the microcomputer 9 shown in FIG. , the sub-aperture may be mechanically selected according to the size of the output image.

In addition, there are various methods for magnification conversion today, both analog and digital (see Japanese Patent Publication No. 44-23651 and Japanese Patent Publication No. 60-37464), and it is possible to adopt these methods as appropriate. can.

<Effects of the Invention> As is clear from the above description, according to the present invention, the spatial frequency component of the image signal to be emphasized is varied according to the size of the output image, so that the size of the output image can be changed. It is possible to obtain a width of edge enhancement corresponding to the width of the edge enhancement, and to obtain a visually pleasing output image with high sharpness.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a sharpness processing section of an image scanning and recording apparatus according to an embodiment of the present invention, and FIG. 2 is an operational flowchart of the embodiment. 3 to 6 are diagrams for explaining the present invention, in which FIG. 3 is a schematic diagram of an aperture used for photoelectric scanning, FIG. 4 is a waveform diagram of each signal related to sharpness processing, and FIG. The figure is a spatial frequency characteristic diagram of each of the signals, and FIG. 6 is a schematic diagram showing an example of a pixel area used in the digital processing method. 1... Line memory group 2... Averaging processing unit 8...
・Selection circuit 9...Microcomputer IO・
...Manuscript size input section 11...Trimming range input section 12...Reproduction magnification input section Figure 2 Applicant Dainippon Screen Mfg. Co., Ltd. Agent Patent attorney Tsutomu Sugitani Figure Figure Figure

Claims (2)

[Claims] (1) Find the size of the output image based on the specified conditions, estimate the standard viewing distance according to the size of this output image, and make sure that the output image is sharp when viewed from this viewing distance. 1. A method for enhancing sharpness in an image scanning and recording apparatus, characterized in that a required spatial frequency component of an image signal that strongly contributes to sharpness is determined, and this spatial frequency component is emphasized. (2) In the sharpness enhancement method in an image scanning recording apparatus according to claim (1), the required spatial frequency component of the image signal to be enhanced is 12 cpd (cycle per d
A sharpness enhancement method in an image scanning recording device that is a spatial frequency component near egree.